The 1,2-product is drawn below.
As for hard/soft theory it goes like this:
All nucleophiles and electrophiles can be ordered by their relative hardness. This has to do with how diffuse their charge (partial or full charge) is on the molecule or atom. For example, H+ would be a very hard cation because all of the charge is centered on one small atom. In contrast, the cesium cation would be relatively soft because the atom is very large but bears the same charge. In multi-atom systems you might also need to consider other resonance structures; for example, the allyl cation is relatively soft because the positive charge is spread out across the three carbon atoms that make up the allyl group. Nucleophiles can be classified in the same way. For example, alkoxides are hard nucleophiles while thiolates are relatively soft. In the case of an enone, there are two potentially electrophilic sites: the carbonyl carbon and the beta-carbon. The carbonyl carbon is the "harder" of the two electrophilic sites because it bears most of the positive charge based on the relative contributions of the different resonance structures.
So how all that relates to reactivity is quite simple: hard nucleophiles prefer to react with hard electrophiles, soft nucleophiles prefer to react with soft electrophiles. In the case of this question, a hard nucleophile (organomagnesium) is used so you would predict addition to occur at the harder electrophilic site, the carbonyl carbon). If you were to use a different organometallic reagent like an organocuprate, then you would get addition to the enone. One explanation of this is that organocuprates are softer nucleophiles than organomagnesium compounds. (In reality it is likely quite a bit more complicated than that, but that is the pertinent explanation for this case)
EDIT: forgot to attach the picture!